Guidance calibrating Tascam MS-16 for ATR tape

Hi helpful friends! I modified my Tascam MS-16 amp cards and speed to be 15ips. It’s now also modified to have additional bias current available.

So I bought a reel of ATR master tape.

My calibration tape is 250nWb/meter.

If I want to get the most out the headroom for this ATR tape, how will I change my calibration process?
1.)Do I just calibrate playback to be -6dB on the meters??

2.) Then when calibrating the record levels, do I increase the voltage of my signal generator? What voltage am I aiming for?

3.) someone said there’s a switch under the bottom panel of my Tascam Amp card section that selects +8, -10, etc. This complicated things for me. I’ll dig into the manual and see about this but if you can give me the nutshell of how this changes the calibration process, thank you.

4.) I’m using a Tascam M-3500 board. Can I just use the test tones from the board and set the VU meters on the tape machine for -6,8,9(?)…?

I just want to say thanks, I’d be lost without this community.
1) If you want your operating level to be 500nWb/m then yes.

2) Keep the signal generator at the operating level of the machine - probably +4dBu although possibly -10dBV.

3) Keep things consistent - if your desk uses +4dBu then set the recorder to that.

4) Yes, that's what the tone generator in the desk is for. I'd record tones at all the frequencies that you generator can produce. While 1kHz, 10khz and 100Hz are standard, it is very helpful to have the additional 40Hz and 16kHz too.
I'm not convinced that setting your machine up for extreme elevated levels will gain much in the way of performance, especially if you intend to use dbx I NR. Regardless, 3db hotter than 250nWb/meter is as high as I'd attempt. You changed your cards but not your heads and they might not tolerate anything higher.
While I was adjusting BIAS for channel 12 of 16, the VU meter suddenly dropped from +2 to -20. Now it won’t go above -20.

Im worried I just destroyed my BIAS amplifier module. I’m blaming myself for not using a plastic screwdriver.

I think I just took 5 years off my life from stress.
I can’t even calm down enough to come up with a question.

Pink noise recorded through the channel comes out flat. Tell me I didn’t just ruin my amp card.

PS There is nobody within 500 miles of me who’s willing to calibrate this machine for me for payment. I’m stuck doing it myself.
Last edited:
Hold on, this means that 10khz now comes out as -20 VU. Perhaps I broke a component I installed from the 30 IPS to 15 IPS modification.
I’m really thinking I have to buy a new amp card now. Bummer. Is there a replaceable component responsible for high frequencies in the amp card?
With a bit of luck you have just broken the trimmer. There are a number of components that could be problems - however, your saving grace is that you have a number of identical channels - so you can measure these and soon work out if something is different. You have the service manual I assume, so just do the bias alignment using the levels in that document. In my experience, if you adjust bias and the levels outside of the manual, you need to use more tones to ensure you are creating the linear section of the curve, so your max and mins are in that straight line portion. You might also need a scope to make sure the levels are not distorting the waveform, so you might have two conflicting adjustments - you squeeze a bit more level, but it starts to distort or over saturate. Probably best to set up a spreadsheet to record the details of the frequency, the input and output levels and the point distortion starts - if you plot these as input against output you are looking for a linear straight line that runs at 45 degrees signifying the output from tape matches the rising input level. Realistically the ideal is the noise at the bottom being manageable and within your own limit. To be honest I'm not sure chasing maximum level works that well as the usual increase in input gain to get the level can introduce more noise. I've always thought of it like a big engine in a car vs a highly tuned smaller one. Pushing that 250nWb limit might be pointless as Rick says, and remember to always check your erase head can actually manage to erase the higher level record head magnetism. I've usually worked on the principle that if ramping up the levels would have improved performance by a worthwhile degree for zero cost - they would have done it in the factory.
Thanks Rob! I’ll grab a trimmer and replace it, maybe that’ll work. But I moved the amp card to channel 16 so it’s not used right now.

Thank you for this information. And I did decide not to change the calibration to +9. I’ll just hit the tape harder to see if I can get farther above the noise floor—no use changing all my settings if I might have to change it all back.

I’ll try to remember to update this thread with my results of replacing the trimmer.